Almost every task used in studying human behavior demands that subjects make categorical responses int eh face of ambiguous or incomplete information. This requirement introduces a decision-making component into many experiments that are designed to measure other aspects of perception or cognition. The strategies subjects adopt to cope with this decision-making component can dramatically and misleadingly influence experimental outcomes if they are not explicitly considered by the investigator. The aim of this proposal is to investigate the role of strategic thinking in behavioral tasks and to investigate specific neural mechanisms associated with decision making and strategic though processes. We propose a series of experiments to investigate the role of the left hemisphere and the prefrontal cortex in hypothesis formation. We believe that subjects have a strong tendency to look for and to posit causal relationships even when none exist. There is evidence that this tendency is supported by structures in the left hemisphere in most individuals. This type of thought process is of real benefit when there is a logical relationship among events but is potentially detrimental if there is no casual relationship exists. We propose a series of studies testing normal subjects, callosotomy patients, and lesion patients that vary the nature of the relationship among elements of the task and examine the presence and nature of hypothesis that are formed. Data from the patient populations will serve to identify cortical regions fundamental to these tasks. We believe that these two sets of studies will have a significant impact on the way that imaging studies and patient studies are evaluated, and will further our understanding of how the brain makes a decision.